Susceptibility-weighted imaging (SWI) is an advanced magnetic resonance imaging (MRI) technique that dramatically improves the detection and characterization of brain hemorrhages. By exploiting the intrinsic magnetic susceptibility differences of various tissues and blood products, SWI produces images with exquisite contrast, enabling radiologists to identify hemorrhages that would otherwise remain invisible on conventional MRI sequences. This article provides a comprehensive overview of the principles behind SWI, its specific mechanisms for detecting brain hemorrhage, its clinical applications, advantages over other imaging modalities, limitations, and emerging technological developments.

Understanding Magnetic Susceptibility and SWI Principles

Magnetic susceptibility refers to the degree to which a material becomes magnetized when placed in an external magnetic field. In biological tissues, variations in susceptibility arise from the presence of paramagnetic, diamagnetic, and ferromagnetic substances. Paramagnetic materials, such as deoxyhemoglobin and hemosiderin, have unpaired electrons and become magnetized in the same direction as the external field, causing local field inhomogeneities. Diamagnetic materials, like water and most proteins, become weakly magnetized opposite to the external field. SWI harnesses these differences by combining magnitude and phase information from a gradient-recalled echo (GRE) sequence, thereby generating images that are exquisitely sensitive to local field distortions. This sensitivity allows SWI to visualize small quantities of blood products, iron deposits, and calcium.

The SWI Pulse Sequence: Magnitude and Phase Data

A standard SWI acquisition uses a high-resolution, three-dimensional (3D) GRE sequence with full flow compensation in all three axes to minimize flow-related artifacts. The sequence acquires both magnitude and phase images. The magnitude image provides standard T2*-weighted contrast, while the phase image encodes the local frequency shifts caused by susceptibility differences. The phase data undergo a high-pass filtering process to remove low-frequency background field inhomogeneities. The resulting filtered phase image is then used to generate a phase mask, typically by inversing the negative phase values and normalizing them to 1. This mask is multiplied multiple times (usually four) into the magnitude image to enhance hypointense signals from paramagnetic substances. The final SWI image presents blood products, veins, and mineral deposits as markedly hypointense (dark) regions against a brighter background, dramatically improving their conspicuity.

Mechanism of Hemorrhage Detection with SWI

Blood Breakdown Products and Their Magnetic Properties

When a brain hemorrhage occurs, extravasated blood undergoes a well-characterized sequence of biochemical changes. Oxygenated hemoglobin, present in fresh blood, is initially diamagnetic. Over time, it converts to deoxyhemoglobin (paramagnetic), then to methemoglobin (paramagnetic), and finally to hemosiderin (strongly paramagnetic). Each of these breakdown products has a distinct magnetic susceptibility profile. Deoxyhemoglobin and hemosiderin create significant local field inhomogeneities that cause rapid T2* signal decay, leading to signal loss on GRE sequences. SWI amplifies these effects, making even microscopic deposits of hemosiderin visible as dark foci.

SWI Signal Characteristics of Hemorrhages

On SWI, acute and subacute hemorrhages (containing deoxyhemoglobin and methemoglobin, respectively) appear as well-defined hypointense regions with variable surrounding hyperintense edema. Chronic hemorrhages, where hemosiderin-laden macrophages persist, exhibit dense, dark signal dropout. The sensitivity of SWI is so high that it can detect microbleeds as small as 5–10 mm in diameter, which are often invisible on conventional T1-weighted, T2-weighted, or standard T2* GRE images. This remarkable sensitivity stems from the multiplicative effect of the phase mask on the magnitude data, which accentuates the susceptibility-induced signal loss.

Clinical Advantages of SWI for Brain Hemorrhage Detection

  • Unparalleled sensitivity to microbleeds: SWI can identify cerebral microbleeds (CMBs) that represent chronic, small-volume hemorrhages, often associated with cerebral amyloid angiopathy, hypertensive microangiopathy, and traumatic brain injury.
  • Superior delineation of hemorrhage margins: The high contrast between the dark hemorrhage and surrounding brain tissue enables precise quantification of hematoma size and extension, aiding surgical planning and monitoring of hemorrhage expansion.
  • Distinction between hemorrhage and calcification: While both appear hypointense on SWI, phase images can differentiate paramagnetic blood products (negative phase shifts) from diamagnetic calcium (positive phase shifts in most settings), an important diagnostic discriminator.
  • Early identification of hemorrhagic transformation: In ischemic stroke, SWI can detect minor hemorrhagic transformation that may contraindicate thrombolytic therapy, thus guiding treatment decisions.
  • Non-contrast venography: SWI inherently highlights deoxygenated blood in veins, providing high-resolution venograms without the need for intravenous contrast agents, which is valuable in assessing venous thrombosis.

Comparative Utility: SWI versus Conventional MRI and CT

Computed tomography (CT) remains the first-line imaging modality for acute intracranial hemorrhage due to its speed and widespread availability. However, CT has limited sensitivity for small bleeds, especially those near the skull base or in the posterior fossa. Conventional MRI sequences, such as T2-weighted imaging and standard T2* GRE, detect larger hemorrhages but often miss microbleeds. SWI substantially outperforms these sequences: studies have shown that SWI can detect up to three to six times more microbleeds than conventional T2* GRE imaging. Compared to CT, SWI avoids ionizing radiation and offers superior soft-tissue contrast, though it requires longer acquisition times and patient cooperation. In many clinical protocols, SWI is now a standard complementary sequence in brain MRI for trauma, hypertension, dementia, and suspect microangiopathy.

Key Clinical Applications

Traumatic Brain Injury and Microhemorrhages

Traumatic brain injury (TBI) frequently results in diffuse axonal injury (DAI), characterized by shear-strain damage to white matter tracts. SWI is exceptionally sensitive for detecting the small petechial hemorrhages typical of DAI. These microbleeds often localize at the gray-white matter junction, corpus callosum, and brainstem. Their presence and distribution correlate with injury severity and predict long-term neurocognitive outcomes. SWI-based grading systems for TBI have been proposed to stratify patients for treatment and rehabilitation. Moreover, SWI can differentiate traumatic microbleeds from other causes (e.g., hypertensive) by their characteristic distribution along injury-prone areas.

Stroke and Intracranial Hemorrhage

In acute ischemic stroke, SWI helps identify the presence of any hemorrhagic component within the infarcted tissue, which is a critical contraindication to intravenous thrombolysis. The so-called "SWI susceptibility vessel sign" can also detect acute thrombus in occluded arteries, aiding in the diagnosis and localization of large-vessel occlusion. In patients with spontaneous intracerebral hemorrhage (ICH), SWI reveals the full extent of bleeding and can identify associated microbleeds that indicate an underlying vasculopathy (e.g., amyloid angiopathy). The pattern of microbleed distribution—lobar versus deep—helps distinguish cerebral amyloid angiopathy from hypertensive microangiopathy, which guides subsequent management and prognostic discussions.

Cerebral Amyloid Angiopathy and Neurodegenerative Disorders

Cerebral amyloid angiopathy (CAA) is characterized by the deposition of amyloid beta in small- to medium-sized blood vessels, leading to recurrent lobar hemorrhages and microbleeds. SWI is the imaging modality of choice for diagnosing CAA according to the Boston criteria, as it detects the characteristic multiple, strictly lobar microbleeds and cortical superficial siderosis. In neurodegenerative diseases such as Alzheimer disease and Parkinson disease, SWI can quantify iron deposition in deep brain nuclei (e.g., substantia nigra, basal ganglia), potentially serving as a biomarker for disease progression. The ability to map iron content noninvasively opens new avenues for research into the role of iron in neurodegeneration.

Limitations and Artifacts in SWI

Despite its advantages, SWI has several limitations. The technique is sensitive to motion, especially in uncooperative patients or those with involuntary movements, leading to image degradation. Acquisition times are relatively long (typically 4–8 minutes for a whole-brain 3D sequence), which can be problematic for acute settings. SWI is also prone to susceptibility artifacts from metallic implants, dental hardware, and surgical clips, which may obscure adjacent brain regions. Differentiating between hemorrhage and other paramagnetic substances such as air, certain calcifications, or ferritin deposits can sometimes be challenging, although phase information often resolves this. Furthermore, SWI does not provide direct information about tissue viability or blood-brain barrier integrity, necessitating correlation with other sequences like diffusion-weighted imaging (DWI) or perfusion imaging. Finally, the interpretation of SWI requires experience, as subtle microbleeds must be distinguished from small vessels or flow voids.

Future Directions and Technological Advances

Recent innovations aim to overcome current limitations and extend SWI's capabilities. Ultra-high-field MRI (7 Tesla and above) provides even higher susceptibility sensitivity and spatial resolution, enabling visualization of microbleeds and subtle iron deposition that are imperceptible at 3 T. However, increased field strength also amplifies artifacts, requiring advanced shimming and specialized pulse sequences. Quantitative susceptibility mapping (QSM) is a post-processing technique that solves the magnetic field inversion problem to produce voxel-wise quantification of tissue susceptibility, thereby distinguishing iron from calcium and enabling absolute measurement of magnetic content. QSM is increasingly used in research and is beginning to be translated into clinical applications for hemorrhage quantification and differentiation. Artificial intelligence (AI) and deep learning approaches have been developed to automatically detect microbleeds on SWI, reduce motion artifacts, and accelerate acquisition times through undersampling and reconstruction. These tools promise to improve diagnostic efficiency and reproducibility. Additionally, susceptibility-weighted angiography (SWAN) and other vendor-specific variations continue to refine the technique. As SWI becomes more widely available and standardized, its role in clinical neurology and neurosurgery will continue to expand.

Conclusion

Susceptibility-weighted imaging has fundamentally transformed the detection of brain hemorrhages, providing an unparalleled ability to identify even the smallest microbleeds and characterize the full extent of hemorrhagic lesions. By exploiting the magnetic susceptibility of blood breakdown products, SWI enhances contrast far beyond what is achievable with conventional MRI sequences. Its clinical applications span traumatic brain injury, stroke, cerebral amyloid angiopathy, and neurodegeneration. While limitations such as motion sensitivity and long acquisition times remain, ongoing advances in ultra-high-field MRI, quantitative susceptibility mapping, and artificial intelligence are poised to further enhance its utility. Incorporating SWI into routine brain MRI protocols is essential for comprehensive evaluation of patients with suspected intracranial hemorrhage, ultimately improving diagnostic accuracy and patient outcomes.

For further reading, refer to: Haacke et al., "Susceptibility-weighted imaging: technical aspects and clinical applications, part 1," American Journal of Neuroradiology; Radiology Key overview of SWI; Mittl et al., "SWI in traumatic brain injury: detection of microbleeds and prediction of outcome," Radiology; Greenberg et al., "Cerebral microbleeds and MRI: a critical review," Stroke; and Straub et al., "Quantitative susceptibility mapping: an emerging tool in MR imaging of the brain," Journal of Magnetic Resonance Imaging.